A qualitative analysis of the quality of social and marital support for PTSD victims

Journal Pre-proof A qualitative analysis of the quality of social and marital support for PTSD victims Catherine Fredette Ph.D candidate Elias Rizkallah Ph.D Ghassan ´ El-Baalbaki Ph.D Stephane Guay Ph.D

PII:

S2468-7499(19)30070-5

DOI:

https://doi.org/doi:10.1016/j.ejtd.2019.100134

Reference:

EJTD 100134

To appear in:

European Journal of Trauma & Dissociation

Received Date:

25 April 2019

Revised Date:

8 August 2019

Accepted Date:

19 September 2019

Please cite this article as: Fredette C, Rizkallah E, El-Baalbaki G, Guay S, A qualitative analysis of the quality of social and marital support for PTSD victims, European Journal of Trauma and Dissociation (2019), doi: https://doi.org/10.1016/j.ejtd.2019.100134

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier.

Running head: A QUALITATIVE ANALYSIS OF SOCIAL AND MARITAL SUPPORT FOR PTSD A qualitative analysis of the quality of social and marital support for PTSD victims Authors Catherine Fredette1, Ph.D candidate [email protected]

oo

[email protected]

f

Elias Rizkallah1, Ph.D

pr

Ghassan El-Baalbaki1, Ph.D

e-

[email protected]

Pr

Stéphane Guay2,3, Ph.D

[email protected] Université du Québec à Montréal :

al

1

2

Université de Montréal :

Jo u

rn

405 Rue Sainte-Catherine Est, Montréal, QC H2L 2C4,

2900 Boulevard Edouard-Montpetit, Montréal, QC H3T 1J4, 3

Institut Universitaire en Santé Mentale de Montréal : 7401 Rue Hochelaga, Montréal, QC H1N 3M5

Abstract In recent years, many authors have written on the importance of analyzing social support when it comes to understanding PTSD patients. However, the effects of important aspects of the social support construct itself, such as the source of support and its underlying psychosocial processes, remain relatively unexamined. Using a qualitative design, this study sought to explore social

Page 1 of 32

2

oo

f

support processes between PTSD patients and their relatives. More specifically, thematic and conversational analysis were performed on filmed interactions between 48 PTSD patients and an accompanying person (AP) of their choice. Results indicated that female PTSD patients generally presented a higher frequency of negative interactions than male diagnosed with PTSD . Moreover, couples generally experienced more negative social support compared to non-couples. Finally, it was found that the length of the relationship as well as the onset of the PTSD (before or after the beginning of the relationship) influenced the quality of the interactions for couples in this sample. To our knowledge, this study is one of the few qualitative research that has been made on interactions between PTSD patients and their relatives. This study also offers an innovative point of view on the social processes underlying marital and social support, such as the impact of the length and moment of apparition of romantic relationships on the quality of support offered. Keywords: Qualitative analysis; Positive social support; Negative social support; Marital support; Gender; Conversational analysis; Perceived support; Received support

pr

Introduction

e-

In recent years, the relationships between social support and posttraumatic stress disorder (PTSD) onset, severity and recovery following psychotherapy have been studied extensively

Pr

(Fredette, El-Baalbaki, Rikzallah, Palardy & Guay., 2016; Guay et al., 2011; Ozer, Best, Lipsey, & Weiss, 2003; Brewin, Andrews, & Valentine, 2000). More specifically, the lack of support

al

was shown to be a strong predictor of the development and maintenance of PTSD symptoms

rn

(Ozer, Best, Lipsey, & Weiss, 2003; Brewin, Andrews, & Valentine, 2000). When looking at specific factors believed to influence the link between social support and PTSD, the scientific

Jo u

literature has looked at gender, marital functioning and the source of support, amongst others. Crevier, Marchand, Nachar, and Guay (2014) found that gender acted as a moderator of the links between interactional behaviors (perceived as either negative or positive social support) of individuals with PTSD and their concurrent depressive symptoms. Another study by Crevier, Marchand, Nachar, and Guay (2015) also indicated that intimate partners (when compared to family and friends) were less involved and more often displayed signs of dysphoria and distress when interacting with people suffering from PTSD, which could influence on the quality of the support offered. Allen, Knopp, Rhoades, Stanley, and Markman (2018) found that specific

Page 2 of 32

3

symptoms of PTSD (e.g. emotional numbing) were linked to lower levels of marital functioning and higher marital conflicts, which has an influence on social support. Indeed, marital satisfaction was recently found to be associated with both positive and negative measures of social support (Fredette, El-Baalbaki, Rizkallah,& Guay, 2016), where higher marital satisfaction is associated with higher positive social support and lower negative social support. A systematic

f

review of the literature also examined the link between marital and social support on the efficacy

oo

of CBT for PTSD and found that cognitive-behavioral couples therapy (CBCT) for PTSD could positively influence marital support (Fredette, El-Baalbaki, Rikzallah, Palardy & Guay., 2016).

pr

These results thus suggest the influence of factors related to gender, the source of support, and

e-

marital interactions on the relation between social support and PTSD symptomatology. Sources of Support

Pr

Social support is derived from the social network of the person and can come from many sources, such as the family, the friends and the intimate partner. It can be described as the quality

al

of the interactions within an individual’s social network (Guay et al., 2011). These different

rn

sources of support might as well have differing effects on the quality of support provided. For example, although the intimate partner is generally considered as the most important source of

Jo u

support (Denkers, 1999; Halford & Bouma, 1997), the impact of living with someone with PTSD in the long-term could affect the quality of the marital relationship and of the support offered. It is then possible that different relationships’ lengths or dynamics might affect the quality of the support offered between intimate partners. The Couple Adaptation to Traumatic Stress (CATS) Model (Nelson Goff & Smith, 2005) inform us about the impact of traumas for couple. It stresses that the survivor’s level of functioning or trauma symptoms will set in motion a systemic response that could result in secondary traumatic stress symptoms in the partner. Symptoms of

Page 3 of 32

4

secondary trauma in the partner may then intensify symptoms of primary trauma in the spouse. Moreover, the CATS Model proposes that adaptation to traumatic stress is dependent on the systemic interaction of three primary concepts: Individual level of functioning, predisposing factors and resources, and couple functioning. Based on this model, couples who were together for a long time before the traumatic event might provide a different type of support compared to

f

couples who met shortly after the traumatic event and are still at an early stage of their

oo

relationships. Indeed, more severe pre-existing problems in long-term relationships (predisposing factors) and high stress associated with the impact of the trauma might combine and reduce the

pr

quality of support offered to PTSD patients. In other words, couples whose level of resources

e-

was low before the trauma (e.g. lack of positive coping strategies) might combine with the stress associated with the trauma to reduce adaptation. Moreover, it is possible that friends and

Pr

relatives provide more positive social support in general as the frequency of their encounters with PTSD patients is lower or that they are less emotionally involved in patients’ everyday life.

al

The attachment system is also believed to influence traumatic experiences and adjustment

rn

(Marshall & Frazier, 2018). For example, individuals with insecure or avoidant attachment orientations might rely on negative coping strategies, like exaggerating the seriousness of the

Jo u

trauma (insecure) or downplaying their distress (avoidant) because they are typically unable to obtain the support and security feeling that they need (Marshall & Frazier, 2018). The presence of a family member or a friend with whom the attachment orientation might be more healthy could potentially reduce the impact of a negative early attachment orientation. However, these hypotheses remain to be tested. Type of Support

Page 4 of 32

5

Another important aspect when considering social support is the type of support provided. Indeed, social support providers can unintentionally be unhelpful and negative in their attempt to be supportive (e.g. being critical, blaming, avoiding; Scarpa, Haden, & Hurley, 2006). According to Ray (1992), the type of support has a different impact on anxiety, whereas positive social support is positively correlated with anxiety while negative social support has a negative

f

correlation. Positive social support refers to supporting behaviors such as listening to the other

oo

person, validating, and proposing positive solutions whereas negative social support is characterized by non-supporting behaviors such as criticism, dominance and rejection (Guay,

pr

Marchand, & O’Connor, 2003; Pizzamiglio, Julien, Parent & Chartrand, 2001). been described

e-

as negative social interactions, interpersonal frictions, interpersonal stress and social constraints (Guay et al, 2011). The type of support (negative versus positive) is thus determined by the

Pr

nature of the behaviors and the content of the message of the social support provider. The impact of negative social support on PTSD symptoms severity more specifically have been described by

al

a few authors (Fredette, El-Baalbaki, Rizkallah, and Guay, 2018; Guay et al., 2011). Moreover,

rn

Nickerson and colleagues (2017) found that PTSD symptoms severity predicted subsequent increases in negative social support as well as decreases in perceived positive social support

Jo u

across time. These studies thus point to the divergent impact of the type of social support in relation to PTSD. However, only a few studies have included a measure of negative social support, and factors related to the different sources of support (e.g. gender, type of relationships) in relation to the provision of negative social support are not well known. Methods for Assessing Social Support When assessing social support, most studies analyze perceived social support, which refers to one's perception of the quality of his/her social support. Few authors have assessed

Page 5 of 32

6

received social support, which refers to provided social support to the patient. Lehoux, Guay, Chartrand and Julien (2007) adapted an observational coding system (Système de codification des interactions de soutien - Macroscopique; SCIS-M, by Pizzamiglio, Julien, Parent & Chartrand, 2001) for people suffering from PTSD, which was subsequently used in many studies in order to assess received social support (Crevier et al., 2015; Crevier et al., 2014; Nadim,

f

Lavoie, Marchand, O'Connor, & Guay, 2014). Qualitative methodologies, like thematic and

oo

conversational analyses, could also inform us on the difference between perceived and received social support. Indeed, the analysis of discourse content could assess the perception of social

pr

support whereas conversation analyses of interactions could provide insight into the quality of

e-

social support offered to PTSD patients. Yet, the authors of this article are not aware of other studies using such methodologies to analyze perceived and received social support in the field of

Pr

traumatology.

Objectives

al

Although a considerable amount of research was made on the link between PTSD and

rn

social support, the effects of important aspects of the social support construct itself, such as the source of support and its underlying psychosocial processes (such as dyads’ dynamics or

Jo u

interactions), remain largely unexamined. The objective of the present study is thus exploratory in nature and attempts to qualitatively describe the type and quality of social and marital support provided and perceived during an interaction between a PTSD patient and a AP of their choice. Methodology Participants Participants in this study were 48 dyads composed of a person with a main diagnosis of PTSD (PDP) and a person (accompanying person; AP) of their choice, which could either be an intimate partner (spouse/loved one), or a non-intimate partner like a family member or a friend.

Page 6 of 32

7

Inclusion criteria were the following: having a main diagnosis of PTSD, being aged between 18 and 65, being able to write and speak in either French or English, and having an AP who is willing to participate. The Clinician-Administered PTSD Scale (CAPS; Blake et al., 1990)1 was used to diagnose the presence of PTSD. People were excluded for the following reasons: being psychotic, having a somatic symptom and related disorders, having an eating disorders, an organic disorder

f

or a disorder induced by medication. Moreover, participants with a history of marital violence were

oo

excluded in order to prevent a dangerous escalation of conflicts during the experiment. For a description of participants in this study, see table 1.

pr

Data Collection and Analysis

e-

This study was carried out at the Centre d'étude sur le Trauma of the Institut Universitaire en Santé Mentale de Montréal (IUSMM) between 2007 and 2010. The data used are part of a

Pr

larger study (secondary data) that aimed to evaluate the physiological correlates of anxiety during a filmed interaction between two people. Participants (PDP and AP) were first asked to

al

enter the room and to have a neutral discussion during approximately 10 minutes. Then, they

rn

were asked to have an active conversation for approximately 15 minutes. For the neutral discussion, the instruction was to avoid themes that normally create conflicts or disagreements.

Jo u

As for the active conversation, the following instructions were transmitted to the participants: “Talk about the trauma and its impacts, about how you help each other out and how you imagine the future ”. Only the active conversations were analyzed for the purpose of this study. First, a complete transcription of the discussions was made with the software program Inqscribe. In

1

The original version presents good psychometric properties. It presents good internal consistency coefficients (varying from α = 0.73 to 0.85) as well good inter-rater reliability (r = 0.92 to 0.99 ) for all three categories of symptoms (Blake et al., 1990). Moreover, the CAPS has good correlations with the Mississippi scale (r = 0.70; Keane, Caddell, & Taylor, 1988) and PK sub-section of the MMPI (0,84; Keane, Malloy, & Fairbank, 1984), which demonstrates good convergent validity (Blake et al., 1990).

Page 7 of 32

8

order to make transcriptions as uniform as possible, preprocessing was necessary. A standardized sheet was thus created in order to standardize the transcription process as much as possible (see appendix A). In order to apply a conversation analysis, the method of transcription followed, albeit in a less strict manner, was the Jefferson method (Jefferson, 2004), with the aim of staying as close as possible to the speech of the participants (e.g. gabbling, hesitations, silence duration,

f

overlapping talk)2. All the coding, thematic and conversational, was conducted with the software

oo

QDA Miner. First, a thematic analysis was conducted following a sequential coding gait in order to analyze the content of the discourses (Paillé & Mucchielli, 2012; Boyatzis, 1999). More

pr

precisely, open coding was first made on a small sample while staying as close as possible to the

e-

discourses of the participants. Then, a thematic tree was inductively built which comprised many hierarchical levels of codes and categories. The small sample consisted of 12 videos that were

Pr

chosen in order to have the most varied discourse content videos and because they covered all of the different conditions of the study (e.g. couples, non-couples, families, friends, different

al

lengths of the relationships). Then, major revisions were conducted in order to solidify the tree

rn

(merging, renaming, changing the hierarchical levels or suppressing codes in order to have the most exhaustive and non-redundant collection of codes). Codes and categories were then

Jo u

presented to another researcher in order to approve of their stability, changes were made if necessary. Codes were then applied to the rest of the videos in the most inductive and iterative manner possible. Moreover, an effort was made in order to understand social support in terms of positive/negative perceptions. A conversational analysis of the transcript was also carried on all active conversations in order to analyse the interactions found within the discourses of participants (Hutchby & Wooffitt, 2008). The construction of conversational codes was more

2

It is of note that language used by participants was wither French or English. For the purpose of the readability of the article, verbatim quotes are going to be presented in translated English with only relevant transcription marks.

Page 8 of 32

9

deductive in nature. Indeed, analysis was based on Hutchby and Wooffitt's guidelines while looking for the formal properties (verbal and non-verbal) of discourses. Similar to the thematic analysis, dynamics were grouped under neutral, positive or negative types of interactions and comprised only two levels of categories and codes. See table 2 and 3 for a summary of thematic and conversational codes. This study was conducted under a post-positivist posture in qualitative

f

research (Huberman and Miles, 1994) and followed mostly data management and analysis

Results

oo

methods (Ryan and Bernard, 2000).

pr

Thematic Codes Relating to Social and Marital Support

e-

Perception of support by the PDPs. An analysis of frequency of codes (how frequent a code is present within all of the cases3, expressed in percentage) revealed that approximately

Pr

37.50%4 of the sample naturally reported that they lacked support5 (all sources confounded) whereas only 23.00% of the sample reported that they received some sort of support. More

al

specifically, when the PDP is a woman, the lack of support was reported in 39.00% of all cases,

rn

whereas the presence of support was discussed in 28.00% of all cases. When the PDP is a man, the acknowledgement of support was present in 8.00% of all cases and the absence of support

Jo u

was reported by 33.00% of all cases. Some differences also emerged. Male PDPs never reported in this study that they were not supported by their AP, that they lacked mutual support or that they would need more support, contrary to female PDPs. Moreover, male PDPs reported more lack of support from their employer than female PDPs. These results might thus indicate that 3

One case refers to a single conversation (15 mins) between a PDP and his/her AP. Analysis could be performed on different granularity levels: among cases, on the paragraph level, etc. Then, they could be crossed with different types of comparisons: The speaker (PDP or AP), type of dyad (couple or non-couple), gender, etc. 4 Frequency analyses only refer to the percentage of a code being present on a specific granularity crossed or not with a type of comparison. In this context, formally the absence of a code means it was not expressed in the verbatims and thus its absence is not included in the analysis mainly because of its polysemy, among other reasons. 5 Italicized text represent the name of codes

Page 9 of 32

10

men did not perceive the presence of positive social support compared to women, who rather perceived a greater lack of social support. However, not all PDP expressed themselves about the presence or absence of social support, and thus we cannot conclude about their perceptions. Co-occurrence analyses (Jaccard index6). An analysis of co-occurrences of codes was made in order to see which codes co-occurred the most within a same conversation (case). When

f

looking at the quality of the social support provided by the intimate partner, results indicated that

oo

when the PDP complains of not being supported by intimate partner, the most co-occurring codes (0.33) were PDP relies excessively on the intimate partner, PDP refuses help, little or bad

pr

communication, and others don't see my distress. PDP says “You were not helping me, you were

e-

not helping me at all. You were there…we did not talk (name of the AP). We did not talk anymore…” (video 38, not being supported by intimate partner; little or bad communication).

Pr

The pattern seemed to differ when the PDP complains of not being supported by other people in general. PDP says “ […]the hospital were very bad to me, at that time…and a big chance she was

al

there to support me {referring to lawyer}because nobody did” (video 46, not being supported by

rn

other). Codes that co-occurred the most were not supported by family (0.44), feeling good alone (0.33), needing to talk (0.30), others don't talk about the trauma (0.25) and others don't see

Jo u

distress (0.25). As for when the PDPs reported that they not supported by families, the most cooccurring codes were Not supported by other (0.44), and others don't see distress (0.30). PDP says “Its* very important for me you know to have you and I need support…I don’t have that much you know my parents [theyre], AP says [hmm hmm] PDP says they avoid that all the time my brother I don’t talk really him…heu with him about that..” (video 46, not being supported by

6

A similarity measure between codes on a certain level of granularity (e.g. cases, paragraph). A comparison is computed between where (here case) both codes appear to where one of them appears (same weight given to both situation

Page 10 of 32

11

family). Barriers to support were also observed, that are either caused by the PDPs (e.g. not wanting anybody around them) or due to other people (e.g. having to hide the trauma). A regrouping was made with all codes relating to barriers to support and submitted to a within-case co-occurrence analysis. The most co-occurring code were hide his/her emotions (0.38) and PDP female talking (0.32). It thus appears that barriers to support were mainly expressed by female

f

PDPs. For a summary of results, see table 4.

oo

Analysis of Conversational Codes

The conversational codes were analyzed according to variables that may influence social

pr

and marital support, these are: gender of PDPs, type of relationship (couple versus non-couple),

e-

duration and moment of apparition of the relationship (according to the diagnostic of PTSD). Gender. Like it was reported above, dyads with a male PDP reported in a lesser

Pr

percentage of cases (8.00% males instead of 28.00% females) that they received support, compared to when the PDP is a female. However, an analysis of frequency of conversational

al

codes demonstrated that dyads within which the male is a PDP generally presented less negative

rn

interactions, as demonstrated by no instance of requests that reflected a lack of support (e.g. asking to be listened to), compared to its presence in 18.90 % of cases when the PDP is a female.

Jo u

Male PDPs also presented less negative non-verbal signs (e.g. irritability; 9.10% versus 16.20% for female participants) and less interactions that reflected a negative dynamic (e.g. topic conflict, wanting to get your point across anyway, domination of one partner; 63.60% versus 86.50% for female participants). PDP says I don’t want them to come to my side (talking about a specific type of people on the other side of the street). AP says You, you wont go to their side. That’s what I am telling you, that’s what I am telling you. PDP says This is what I am saying,

Page 11 of 32

12

it’s not like that, it’s not what I want, I don’t want to see them, can I ? AP says What are you going to do in order to not see them ? PDP says Stay at home. AP says No, you will go the other side of the street (video 3, domination). However, dyads in which the PDP is a male swore a lot more (58.30% versus 38.90% for female participants). Although dyads in which the PDPs were males presented less negative interactions,

f

no important differences were found in terms of codes that reflect a positive non-verbal

oo

communication (e.g. laughing), gestures/expressions that reflected positive social support (e.g. offer his/her help) and negative tonalities (ex. exasperated tone of voice) across gender.

pr

Types of Relationship. An analysis of frequencies of the interactions between dyads

e-

seemed to demonstrate that dyads formed by couples (relative frequency)7 presented more negative interactions in general than dyads formed by family or friends. Indeed, there were more

Pr

instances in which the couple dyads speak with a negative tone of voice (e.g. sarcastic tone of voice, 44.00% versus 0.00%), as well as more disagreements (81.60% versus 50.00%), more

al

demands that reflect a lack of support (e.g. asking the romantic partner to stop or asking the

rn

romantic partner to be listened to; 15.80 % versus 10.00%), more negative non-verbal behaviors (e.g. irritability, hostile tone of voice, having a strong reaction to the other person's discourse;

Jo u

18.40 % versus 0.00%) and more non-supporting communication (e.g. criticism, invalidation; 52.60 % versus 10.00%). AP says “Me hum…despite your your…character…which is a little bit hum…which is a little tougher to deal with, hum…I feel more important to you which is ok for me. PDP says @8Ah fuck wow a positive side [wouh hou!]@” (video 13, sarcastic tone of voice). PDP says “I said it is a fucking step AP says A step? PDP says It is a difficult level if you prefer, fuck I am going to explain it to you in a language that you can understand…”(video 2,

7 8

Here relative frequency was applied to balance the over re-presentation of couple dyads Transcription mark indicating a sarcastic tone of voice

Page 12 of 32

13

hostile tone of voice). See Appendex B for quotes in English and in their original and lengthy forms. Interactions between non-couples were rather characterized by more instances of gestures/expressions that reflected their attempts to support the PDP (e.g. offer his/her presence, invite the PDP to express his/her emotions; 90.00 % and 78.90 %, respectively). Moreover, there were no instances of a negative tone of voice (e.g. hostile tone of voice) within these dyads.

f

Time of Onset of Relationship versus time of onset of PTSD. Results of analyses of

oo

frequencies seemed to demonstrate that interactions differed whether couples were formed before or after the PDP was diagnosed with PTSD. Couples formed after the traumatic events

pr

tended to have a dynamic that was influenced by the presence of the disorder. For approximately

e-

65.00% of couples who were formed after the trauma was present, the AP seemed to be an active actor in the PDP's healing and well-being. For example, the PDPs reported being more secure

Pr

and at ease when the intimate partner was present. Moreover, it seemed that the spouse helped the PDP with PTSD symptoms such as sleeping problems (sleep is easier with AP). PDP says:”

al

Since I’m with you I…found my sleep back…it is still troubled but at least it’s there” (video 21,

rn

Sleep is easier with AP). The AP also seemed to help the PDP by providing a more stable life (More stability with AP), or by making the PDPs interact with new and helping people (create

Jo u

positive meetings for me, more stability with AP). AP says “With me you have a certain stability, you know, I have a regular schedule I am always there at the same time [...] PDP says Yes yes yes. Stability will help also. Yes…” (video 27, More stability with AP). For one couple, the dynamic was one in which the PDP has depended a lot on the presence of the partner in order to feel better, which created conflicts within the couple (to invade the other). PDP “I knew you needed to breathe but I stuck to you, I stayed there, but I knew that you needed more space to breathe...[...] It led to the fact that you rejected me” (video 26, Invade the other). Finally, there is

Page 13 of 32

14

also one case in which the spouse seemed to be emotionally over-involved with the PDP. For example, the new spouse seemed over-protective of the PDP (blows the PDP's nose) and got mad during the filmed interactions when the PDP had to talk about something difficult. PDP says: Well...what else do you want me to do, it happened it happened and it's not my...it's not my fault (cries). AP says: (cross talk at the beginning)[...]Did I say once that it

f

was your fault? Does someone understand better than me ? Hen?. PDP says I feel

oo

nauseous AP says: Yes, I know. Breathe. Sit up right. Brea, Breathe. No, control yourself (name) you can do it I know you can. (video 3)

pr

An analysis of frequency by video/case demonstrated that when the AP seemed emotionally

e-

over-involved, there was a much higher frequency of codes relating to being prescriptive (25 times versus 5 and less) and domination of the conversation by the AP (5 times versus 0),

Pr

compared to other couples formed after the trauma.

For couples who were together before the traumatic event, the following codes were

al

present. The spouses talked about sacrifices that they make, dynamic changes within the

rn

relationship, daily adaptations (e.g. being with the other person on a daily basis, moving and changing habits to help the PDP) and about the fact that they do not know how to help the PDP.

Jo u

AP says “ Because you know when sometimes you tell me to come to the house bla bla bla I am not able to sleep...I ask myself if I do the right thing when I go. It is helpful to go to your house and say everything is ok” (video 8, do not know how to help the PDP). It was then hypothesized that these changes could create more negative interactions for these couples. A look at the results (frequency analyses within all cases according to the development of their relationships) seemed to indicate that couples who were together before the trauma generally presented more negative interactions than couples who were formed after. They did have more negative non-verbal

Page 14 of 32

15

interactions (22.00% versus 0.00%) and demands which demonstrate a lack of support (e.g. PDP asking to being listened to; 18.80% versus 0.00%), which was not found at all in couples who were formed after. Moreover, the tonalities of the interactions were generally more negative (e.g. exasperated tone of voice, sarcastic tone of voice; 46.90% versus 20.00%), there were more cases in which questions were not answered (59.40 versus 40.00) as well as more indications of

f

conflicts between the spouses (e.g. to reiterate your point of view many times: 31.30% versus

oo

0.00 %; argumentation: 6.30% versus 0.00%).

No major differences were found between the couples formed before or after the

pr

traumatic event in terms of positives interactions, which is an important information. It thus

e-

seemed that couples formed before the traumatic event might be more inclined to provide neutral or negative social support only. It is important to note that these results were obtained after a

Pr

case was filtered, which was a couple formed after the traumatic event in which the dynamic reflected an emotionally over-involved spouse. This couple differed greatly from other couples

al

formed after the diagnosis of PTSD in this study (see abstract above).

rn

Duration of the Romantic Relationship. Analyses of frequencies revealed that both medium (5-15 years) and long-term (15 years and longer) relationship couples displayed more

Jo u

negative interactions than short-term couples (5 years and less). Indeed, couples who were together for more than five years displayed a greater amount of negative non-verbal signs (e.g. blaming the other person), demands that reflected a lack of support (e.g. asking the romantic partner to be more attentive; expressing that they need help; ranging from 23.10 – 20.00% compared to 6.70%; five years and less), and more disagreements (ranging from 92.30 – 100.00% versus 60.00% for couples together five years and less). PDP says

Page 15 of 32

16

See thats all the problem, you cant talk about it. AP says: No no. You think thats wrong when (inaudible) a 100km an hour, its* an 100km an hour highway, ok? Thats when you start to panic I dont* know why. You start to panic you always slow down. When you slow down on an 100km you become the accident. You become the danger. You become the danger. Then you ll cause an accident cause you re slowing

f

down on an highway. You got to learn to go with the flow (name), you used to do

oo

it. (video 2, Blaming the other person)

Analyses of frequency also demonstrated that medium-term couples (five-15 years) presented

e-

term couples and 30.00% for long-term couples).

pr

more non-supporting communication (69.20%) compared to other couples (53.30 % for short-

In terms of positive social support, short-term couples displayed more positive non-verbal

Pr

interactions such as laughter and acknowledgement (93.30 % versus 70-76.00%) than medium and long-term couples as well as more gestures/expressions that reflected support (offer his/her

rn

Other Relevant Results

al

help, validation; 86.70 % versus 69-80.00%).

Passive versus active. One purpose of the conversational analysis was to explore if

Jo u

different styles of communication emerged between PDP and AP. During the closing phase of coding, the first author observed that some categories (set of codes) emerged and represented passive and active dynamics within the conversations. For example, codes like asking the other person to take the lead of the conversation or saying that they do not know what to talk about were classified under the category "passive type of communication". PDP says “ So. I don’t know where to being, its* not for me those things (referring to the study). ((sigh)). What do you want me to say?” (video 2, Do not know what to talk about). Other participants rather took an

Page 16 of 32

17

active role in the conversations. The following codes were classified under the category "active type of conversation": engaging the conversation and re-orienting the conversation toward the right topics (see table 3 for more precision). An analysis by variables demonstrated that passive and actives dynamics was only found in dyads formed by intimate partners. Passive actions were taken by both the PDP and the AP while active actions (three instances) were taken by the PDP

f

only. In most but not all cases, passive actions were associated with negative interactions.

oo

Examples of co-occuring conversational codes were reflect lack of support (0.30), to reiterate your point of view (0.27). They were also associated with the following thematic codes negative

pr

communication (0.18) and incomprehension (0.15).

e-

Impacts of communication and comprehension on social support. One subject that stood out in this sample was the importance for the PDPs of being understood. Indeed, PDPs felt

Pr

more support from their relatives when they believed to be understood. A co-occurrence analysis (Jaccard index) revealed a difference in perception in dyads formed by romantic partners. The

al

most co-occurring code to not understood by the AP (PDP speaking) was AP understands the

rn

PDPs (0.40). This difference in perception was generally found in heterosexual couples (8 cases) within which the PDP was mostly a woman (6/8 cases). No instances were found in non-couples.

Jo u

This might indicate that couples had more difficulties communicating their emotions related to PTSD symptoms. This result goes in hand with results presented above, which indicated that couples presented a higher percentage of negative interactions compared to non-couples. In terms of family members, misunderstanding was also reflected by the presences of the following codes: minimization of trauma and bad advice (e.g. return to work). Further analyses of co-occurrences were performed in order to determine if codes reflecting a lack of understanding (all sources confounded) co-occurred within the same case

Page 17 of 32

18

with codes reflecting weak levels of communication (e.g. few or bad communication; 0.21, and not sharing everything; 0.21). These results seemed to indicate that weak levels of communication could lead to misunderstandings. One hypothesis then tested was that bad communication would also affect the quality of the social support provided. Analyses of cooccurrences indicate that codes representing weak levels of communication (few or bad

f

communication, AP not receptive, others don't talk about the trauma) also co-occurred with

oo

PDPs claiming that they are lacking support (0.38), and were associated with both female PDPs (0.34) and male AP (0.33), which generally represent intimate partners in this study. For a

pr

summary of results, see table 5.

e-

Discussion

The main objective of this study was to explore the type and quality of marital and social

Pr

support via thematic and conversations analyses of interactions between PDP and AP. More specifically, the content as well as the form of the conversations were analyzed in order to

al

discover underlying social support processes. We compared different types of relationships

rn

(couples versus non-couples) and other external factors such as lengths of relationships, gender of the dyads, moment of apparition of the trauma relating to the onset of the relationship. Finally,

Jo u

we analyzed the dynamics of the relationships. Thematic Codes

The thematic analysis of extracts related to marital and social support led to the creation of different categories that could be grouped under positive and negative social support. Amongst these categories, we noted that a few of them converge with another validated coding system, The Social Support Interaction Global Coding System (SCIS- M; Lehoux et al., 2007). These categories were: validation, good advices and listening. However, this study led to the

Page 18 of 32

19

detection of many other positive and negative categories. These categories could be useful because as of today, only a few questionnaires or grid analysis make the distinction between positive and negative social support (QSBA; St-Jean Trudel, Guay, Marchand, & O'Connor, 2005; SCIS; Lehoux et al., 2007). Moreover, these categories were extracted from conversations between PDPs and their AP, which were not influenced by a pre-selected categorisation, like it is

f

often the case with quantitative analysis. These new categories could possibly serve as the basis

oo

for the creation of a new social support questionnaire.

As for the presence or absence of support as perceived by PDPs, in this study, the sources

pr

of support seemed to have an influence. Indeed, couples generally presented more instances of

e-

negative interactions. Moreover, when the PDPs perceived lack of support, codes representing specific dynamics between the spouses emerged as being influential (PDPs relies excessively on

Pr

romantic partner, offers no mutual help or refuses help). It was also found that the actions of the PDPs themselves could hinder the quality of perceived social support. In term of their social

al

network, results seemed to indicate that people around the PDP might feel uncomfortable or not

rn

totally aware of the consequences of the trauma on the PDPs. On the other hand, symptoms of PTSD like feeling estranged or detached from people (feeling good alone), can affect the

Jo u

perception of the quality of support received or hinder the quality of interactions with relatives (Ray & Vanstone, 2009). Finally, when looking at lack of support provided by family, a trend seemed to appear. In fact, most codes revolved around the idea that the trauma is not such a big thing (minimization of trauma) and that returning to work would be a great solution for the PDPs. The families in this study did not seem to grasp the impact of the PTSD and its consequences on the PDPs. PDPs thus felt misunderstood (you really have to live it to understand it). In one case, the family refused to talk about the trauma at all. These results thus

Page 19 of 32

20

inform us that perception of support from PDPs seems to be influenced by different categories, depending on the source of support. Indeed, marital support seems to be influenced by the dynamic of the couple whereas support from relatives seems more related to the incomprehension and lack of knowledge of the disorder by relatives. Future studies could thus benefit from assessing the perceptions of PDPs on the quality of their support from multiples

f

sources, as negative and positive perceptions of social support could coexist. Moreover, it would

oo

be interesting to see in future studies if providing information systematically to family and friends would reduce the lack of perceived social support and promote better communication,

pr

which could positively affect PTSD development and severity. According to the social-cognitive

e-

processing (SCP) model of Lepore (2001), healing is facilitated by the act of discussing about the trauma and its impacts with supporting individuals. These exchanges are thought to create a

Pr

secure environment and challenge negative perceptions developed by trauma patients. However, it is possible that not every individual in the social network of the PTSD patient is able to

al

provide that type of help, explaining the differences found according to the sources of support in

rn

this study.

These results also support other studies that found a differing impacts of the source of support.

Jo u

For example, Cox, Bakker and Naifeh (2017) also found that friends and significant other support (but not family) mediated the effect of emotional dysregulation on depressive symptoms for PTSD patients. Stanley and colleagues (2018), in a multisite study conducted with firefighters, found that less severe PTSD symptoms were associated with support from supervisors, coworkers and family/friends. When all source were entered in a single mode, only support from supervisors was significant. These results seem to go hand in hand with our results

Page 20 of 32

21

finding that support provided by employers, as reported by males in our study, might be an important aspect to consider. Conversational Codes Gender. Regarding gender differences, results indicated that although male PDPs in this sample tented to underreport the presence of support compared to female PDPs, the nature of interactions with the AP did not support this observation. Indeed, negative interactions were

f

present in a higher percentage of cases within cases in which the PDP was a woman (compared

oo

to male PDPs). Moreover, it was found that barriers to support were mainly mentioned by female PDPs (e.g. don't want to talk about the trauma). These results support a study by Crevier and

pr

colleagues presented earlier (2014), who found that gender acted as a moderator of the links

e-

between interactional behaviors of individuals with PTSD and their concurrent depressive symptoms. More specifically, they observed that female PDPs were less implicated and less

Pr

likely to propose positive solutions compared with male PDPs. As for men, they were more implicated and less likely to criticize their AP than were woman. Another study by Andrews,

al

Brewin and Rose (2003) also found that both men and women reported similar levels of positive

rn

social support one month after being a victim of a violent crime. However, they found that women reported significantly more negative responses from families and friends. Although they

Jo u

results differ slightly from ours, they point to the importance of analyzing both positive and negative social support, as negative social support was again found to be more prevalent amongst female PTSD patients. On the other hand, in one study, it was found that women tended to seek more emotional support, and found this strategy more helpful compared to males (Solomon, Gelkopf & Bleich, 2005). The act of sharing negative emotions and putting oneself in a vulnerable situation might negatively influence the quality of the interactions and explain why dyads comprised of female PDPs presented more negative social support. However, it is

Page 21 of 32

22

important to note that interactions with other people than the AP could also explain the differences in social support naturally reported by male PDPs in this study. Indeed, it is possible that they generally receive less support from their families or friends (most males in this study came with their spouse), and did not talk about it during the filmed interactions. A closer look at the results also indicated that in this study, male PDPs generally reported lack of support from

f

their employers. In light of these results, it seems that women receive more negative social

oo

support from close relationships whereas men tend to receive negative support from more external relationships, which is an important distinction. Research indicate that women are twice

pr

as likely as men to develop PTSD (Crevier et al., 2014), it is thus possible that more negative

e-

interactions with close relatives partly explain this difference. These results also demonstrate the importance for males of being supported from their employers, which is rarely assessed in

Pr

conventional social support assessments. They also support many studies (Leffler & Dembert, 1999; Regehr et al., 2000; Weiss, Marmar, Metzler, & Ronfeldt, 1995), which demonstrated a

al

link between traumatic distress and support from superiors. In future studies, it could also be

rn

interesting to analyse its relation to the efficacy of treatment for PTSD, as workplace reintegration is an important challenge following the experience of a traumatic event and a major

Jo u

treatment goal. It is important to note that in this study, most females were part of a couple dyad, which were also found to be more represented by negative social support. Reproducing these analyses with more non-couples dyads in which the PTSD sufferer is a female would help clarify what factors seem to influence the higher frequency of negative social support found for females suffering from PTSD. Type of relationship. Results also demonstrated that the type of relationship had an influence on the quality of social support provided to the PDPs. Couples generally presented

Page 22 of 32

23

higher levels of counter-supportive behaviors compared to non-couples. These results support the study by Crevier and colleagues (2015), who found that during the discussions, life partners were less involved and more often displayed signs of dysphoria and distress compared to friends and family. On the other hand, during their interactions, non-couples presented more positive interactions in general as well as less negative interactions. One hypothesis would be most non-

f

couples do not live together, hence, this could prevent the relationship to degrade due to highly

oo

daily stress and then to have a smaller influence on the quality of the support provided. Whisman, Sheldon and Goering (2010) demonstrated that the quality of the support provided by

pr

the spouse was more associated with psychopathology than support provided by friends and

e-

family. However, only few studies investigated the individual effect of different sources of support for PDPs. Laffaye, Cavella, Drescher, and Rosen (2008) found that veteran peers

Pr

provided relatively high level of perceived social support whereas perceived levels of interpersonal resources from spouses were significantly greater than perceived resources from

al

either nonveteran friends or relatives. However, in this study, spouses were also considered as

rn

sources of personal stress, which relates to the results found in this study. Indeed, although more couples interacted well in this study, few dyads interacted particularly negatively and lacked

Jo u

support, which was not found at all in non-couples. Moreover, Scarpa, Haden and Hurley (2006) found that perceived social support from family and especially friends predicted reduced PTSD severity at low level but not high level of victimisation. One hypothesis was that at high level of victimisation, friends would unintentionally provide negative social support. While the severity of PTSD is not provided in this study, there were no instances of negative social support in this sample when support was provided by a friend, which does not support their hypothesis.

Page 23 of 32

24

Romantic relationships. In this study, the majority of dyads were formed by romantic partners (75.00%), therefore, their interactions were closely analyzed. Results indicated that different aspects of the relationship could influence the quality of marital support provided for victims of PTSD. Indeed, it was found that couples formed after the trauma had a dynamic that was greatly influenced by the trouble (AP being a source of well-being for the PDPs). Although

f

for most cases this dynamic resulted in generally less negative interactions, an understanding of

oo

the costs and benefits for each partner should be kept in mind, as this could hinder progress in therapy (Bélanger et al., 2008). Indeed, as the PDP gets better, the dynamic of the couple could

pr

possibly be modified (e.g. partner feeling less needed) and partners would thus have to reunite on

e-

new basis that would be beneficial for both partners.

For couples who were already together before the trauma, results indicate that partners

Pr

generally have to adapt to the consequences of the trouble. While some of the couples presented adequate or positive interactions, they generally had more negative interactions than couples

al

formed after the traumatic event. This result supports a study by Calhoun and colleagues (2002),

rn

who found that after a traumatic event, spouses faced a wide range of stressors linked to their caregiving role, such as dealing with crisis, PTSD symptoms and financial problems, just to

Jo u

name a few. The results in this study also go hand in hand with Gerlock and collegues (2014) study, who interviewed couples in which one partner suffers from PTSD. Like for this sample, they found couples that were greatly in distress and other that were considered non-distressed. Moreover, they noted that couples who were non-distressed reported communicating directly about the trauma with their intimate partner. In this study, it was also found that PDPs who felt misunderstood and who reported less social support also reported problems in communicating with their partner.

Page 24 of 32

25

It was also found that couples who were together for a longer period (5 years and more) had generally more negative interactions than young couples. These observations support studies done on Expressed Emotion, which found that most caregivers living many years with a patient with a psychiatric illness are classified as high in Expressed Emotion (Hooley, Rosen, & Richters, 1995). Expressed emotion reflects the extent to which relatives of psychiatric patients

f

express critical, hostile, or emotionally overinvolved attitudes toward their family member

oo

(Hooley & Teasdale, 1989). These results also support a study by King, Taft, King, Hammond, & Stone (2006), suggesting that interpersonal problems associated with PTSD may have a

Results also indicated that passive and active dynamics were only

e-

model of social support”.

pr

negative impact on the quality and quantity of social support across time, supporting an “erosion

found in couples mainly formed before the traumatic event. Again, the fact that these patterns of

Pr

interactions were only found in couples might reflect the fact that couples are generally more emotionally involved than non-couples, which contribute to the development of specific type of

al

interactions. The work of Gable and colleagues (2004) demonstrated that relationship well-being

rn

was increased when partners responded actively and constructively to disclosure of positive events but not when they responded passively and constructively (Gable, Reis, Impett, & Asher,

Jo u

2004). In this current study, the authors also found that responding in a passive manner (either the PDP or the romantic partner) was generally associated with actions or gestures that reflected a lack of support. This could be a sign that the person wishes to protect himself/herself from the negative emotions created by the conversation (e.g. not wanting to talk about it or not knowing what to say). However, it also contributes to create further negative interactions and affect the quality of the support provided during an interaction. Avoidance of talking about the trauma, which is a typical symptoms of PTSD, thus seems to hinder the quality of social support offered

Page 25 of 32

26

to PDPs. These results could also be explained by the type of network orientation of the PDP, which is a set of attitudes and expectations concerning the usefulness of employing social resources in times of needs, which individuals usually developed from past experiences (Clapp & Beck, 2009).A negative network orientation then could possibly hinder PTSD patients from seeking social support, if relationship’s past experiences were perceived as rejecting or

f

ineffective. These results thus point to the importance for clinicians to stay sensitive to

oo

communication difficulties experienced by patients since the trauma and to teach healthy communication tools when necessary. This could be achieved with the romantic partners,

pr

relatives or access to helping groups. Moreover, no other studies have assessed specific

e-

relationship factors and their link to social support. The authors of this study found that they are an important aspect to consider as they seem to contribute to the development of negative social

Pr

support. As the only study investigating these relationships factors, replication of these results could allow for the discovery of more specific dynamics than those found in this study.

rn

innovative information.

al

Moreover, an analysis of their impact on PTSD symptoms severity would also provide

Clinical Implications

Jo u

These results have clinical implications. Indeed, clinicians should be aware that for a victim suffering from PTSD, healing could be jeopardized by negative marital interactions and support at home. Integrating the partner into therapy could be promising . A systematic review by Fredette and colleagues (2016) demonstrated CBT for PTSD was associated with increased social support. Il also showed that integrating the partner into therapy led to an increase in marital adjustment and satisfaction, which could influence marital support. Psychoeducation on good communication skills and on how to deal with the PTSD symptoms could also be beneficial

Page 26 of 32

27

for the significant other, as they sometimes report that they don't know how to help. A noteworthy, a study by Billette, Guay and Marchand (2008) focused on improving appropriate supportive behaviors for couples within which one person suffers from PTSD. At the end of the treatment, none of the participants presented clinical PTSD and they all reported significant improvement in the quality of their social support. Another study by Monson, Schnurr, Stevens

f

and Guthrie (2004) also tested cognitive-behavioral couple's treatment for PTSD with seven

oo

veterans and their wives. They found that veterans' relationship satisfaction did not change across therapy. In this study, the median relationship length was higher for most couples (median

pr

length of 29 years). More recently, Schnaider, Siiercic, Wanklyn, Suvak, and Monson (2017)

e-

also found that pre treatment marital support was associated with larger decrease in PTSD symptoms severity. These results thus seem to support our study in terms of factors associated

Pr

with the marital relationship and their influence on marital support and PTSD. There also seems to lack measures that would specifically assess marital support, which could be beneficial for

al

clinicians who wish to specifically assess the presence or lack of marital support for their

rn

patients. General (assessing perceived support in general) and specific measures of social support (assessing perceived support as relating to PTSD), as well as measures of social functioning in

Jo u

general (Tyrer et al,, 2005) could also be used for individuals to assess the importance of addressing social support suffering from PTSD symptoms and referring, if necessary. Moreover, because positive and negative social interactions could be affected by maladaptive cognitions following the trauma, the Postraumatic Maladaptive Beliefs Scales (PMBS) by Vogt, Shipherf, and Resick (2012) could be another useful assessing tool. On the other hand, a clinician could make the choice to encourage the patient to reach help from family and friends outside of therapy, for example during exposition sessions, in order to maximize the potential of the

Page 27 of 32

28

exercises. This study also demonstrated that male PDPs reported that they generally lacked support from their employers, which could have an effect on their recovery. This indicates that much progress is needed in order to render information relating to mental disorders more accessible to occupation area that typically engage men. Conclusion

f

To our knowledge, this study is the first to qualitatively assess the content and nature of

oo

interactions between a PTSD PDP and his/her AP, which provided interesting insights into the dynamic of these interactions. The primary objective was to explore factors that could influence

pr

the quality of both perceived and received social support by PDPs. Results indicated that gender,

e-

the nature, length and moment of apparition of the relationships as well as the dynamic (ex. level and type of communication, social processes) between partners could positively or negatively

Pr

affect the quality of social support. More specifically, some results indicate that these factors have more impact on the provision of negative social support than positive social support. These

al

results might be relevant considering the fact that some studies reported that negative social

rn

support influenced both the severity (Guay et al., 2011; Zwiebach, Rhodes & Roemer, 2010; Zoellner et al., 1999) and the efficacy of treatments for PTSD (Fredette, El-Baalbaki, Rizkallah,

Jo u

Palardy, & Guay, 2016).

Finally, this study also has limitations. The sample was composed of only a few dyads were composed of a few non-couples (family, friends) as well as male PDPs, which limits the scope of the the generalizability of the results for other relatives, who were less represented in this study. Moreover, the fact that most women were also part of a relationship make the distinction between type of relationship and gender of the PDP difficult to differentiate, even though those comparisons were made in this study. Also, participation in this study was

Page 28 of 32

29

voluntary, which possibly biased the results toward more positive interactions. Finally, the fact that no formal framing of the conversations were used (as a grid for semi-structured interviews) in this study rendered the thematic analyses heterogeneous. It is of note that these interactions were not meant to be qualitatively analyzed. Indeed, the duration of discussions about social support was not equal amongst dyads and partners were free to talk about it the way they wanted.

References

oo

exploratory study it allows thicker analyses for future studies.

f

However, this also reflects spontaneous discourses of PDPs with AP and hence, for an

Jo u

rn

al

Pr

e-

pr

Allen, E., Knopp, K., Rhoades, G., Stanley, S., & Markman, H. (2018). Between-and withinsubject associations of PTSD symptom clusters and marital functioning in military couples. Journal of Family Psychology, 32(1), 134. Andrews, B., Brewin, C. R., & Rose, S. (2003). Gender, social support, and PTSD in victims of violent crime. Journal of traumatic stress, 16(4), 421-427. Békés, V., Beaulieu-Prévost, D., Guay, S., Belleville, G., & Marchand, A. (2016). Women With PTSD Benefit More From Psychotherapy Than Men. Psychological Trauma: Theory, Research, Practice, and Policy, 8(6), 720. Bélanger, C., Leduc, A. G., Fredman, S., El-Baalbaki, G., & Baucom, D., H. (2008). Couples et troubles anxieux. Intervention et évaluation. In Wright, J., Lussier, Y., & Sabourin, S (Eds), Manuel clinique des psychothérapie de couple, (p.507-539). Québec: Presses de l'Université du Québec. Blake, D., Weathers, F., Nagy, L., Kaloupek, D., Klauminzer, G., Charney, D., & Keane, T. (1990). Clinician-administered PTSD scale (CAPS). Boston (Mass), 7. Blake, D. D., Weathers, F. W., Nagy, L. M., Kaloupek, D. G., Gusman, F. D., Charney, D. S., & Keane, T. M. (1995). The development of a clinician-administered PTSD scale. (PDF) Journal of Traumatic Stress, 8, 75-90. doi: 10.1002/jts.2490080106 Billette, V., Guay, S., & Marchand, A. (2008). Posttraumatic stress disorder and social support in female victims of sexual assault: The impact of spousal involvement on the efficacy of cognitive-behavioral therapy. Behavior modification, 32(6), 876-896. Boyatzis, R. E. (1998). Transforming qualitative information: thematic analysis and code development. Thousand Oaks, CA: Sage Publications. Brewin, C. R., Andrews, B., & Valentine, J. D. (2000). Meta-analysis of risk factors for posttraumatic stress disorder in trauma-exposed adults. Journal of Consulting Clinical Psychology, 68(5), 748-766. Clapp, J. D., & Beck, J. G. (2009). Understanding the relationship between PTSD and social support: The role of negative network orientation. Behaviour research and therapy, 47(3), 237-244.

Page 29 of 32

30

Jo u

rn

al

Pr

e-

pr

oo

f

Cox, D. W., Bakker, A. M., & Naifeh, J. A. (2017). Emotion dysregulation and social support in PTSD and depression: A study of trauma‐ exposed veterans. Journal of traumatic stress, 30(5), 545-549. Crevier, M. G., Marchand, A., Nachar, N., & Guay, S. (2014). Overt social support behaviors: Associations with PTSD, concurrent depressive symptoms and gender. Psychological Trauma: Theory, Research, Practice, and Policy, 6(5), 519. Crevier, M. G., Marchand, A., Nachar, N., & Guay, S. (2015). Symptoms among partners, family, and friends of individuals with posttraumatic stress disorder: associations with social support behaviors, gender, and relationship status. Journal of Aggression, Maltreatment & Trauma, 24(8), 876-896. Denkers, A. (1999). Factors Affecting Support After Criminal Victimization: Needed et Received Support From the Partner, the Social Network, et Distant Support Providers. The Journal of Social Psychology, 139(2), 191-201. doi: 10.1080/00224549909598373 Fivecoat, H. C., Tomlinson, J. M., Aron, A., & Caprariello, P. A. (2014). Partner support for individual self-expansion opportunities Effects on relationship satisfaction in long-term couples. Journal of Social and Personal Relationships, 0265407514533767. Fredette, C., El-Baalbaki, G., Palardy, V., Rizkallah, E., & Guay, S. (2016). Social support and cognitive–behavioral therapy for posttraumatic stress disorder: A systematic review. Traumatology, 22(2), 131. Fredette, C., El-Baalbaki, G., Rizkallah, E., & Guay., S. (2018). Impacts of differing dimensions of social and marital support on PTSD symptoms severity. Submitted for publication. Gable, S. L., Reis, H. T., Impett, E. A., & Asher, E. R. (2004). What do you do when things go right? The intrapersonal and interpersonal benefits of sharing positive events. Journal of personality and social psychology, 87(2), 228. Gerlock, A. A., Grimesey, J., & Sayre, G. (2014). Military-related posttraumatic stress disorder and intimate relationship behaviors: a developing dyadic relationship model. Journal of marital and family therapy, 40(3), 344-356. Guay, S., Beaulieu-Prévost, D., Beaudoin, C., St-Jean-Trudel, É., Nachar, N., Marchand, A., & O'Connor, K. P. (2011). How do social interactions with a significant other affect PTSD symptoms? An empirical investigation with a clinical sample. Journal of Aggression, Maltreatment & Trauma, 20(3), 280-303. doi: 10.1080/10926771.2011.562478 Halford, W. K., & Bouma, R. (1997). Individual psychopathology et marital distress. In W. K. Halford & H. J. Markman (Eds.), Clinical handbook of marriage and couples interventions (pp. 291-321). Hoboken, NJ, US: John Wiley & Sons Inc. Hooley, J. M., Rosen, L. R., & Richters, J. E. (1995). Expressed emotion: Toward clarification of a critical construct. In: Miller G.A. (eds), The Behavioral High-Risk Paradigm in Psychopathology (pp.88-120). Series in Psychopathology. Springer, New York, NY Hooley, J. M., & Teasdale, J. D. (1989). Predictors of relapse in unipolar depressives: expressed emotion, marital distress, and perceived criticism. Journal of abnormal psychology, 98(3), 229. Hutchby, I., & Wooffitt, R. (2008). Conversation analysis. Polity Jefferson, G. (2004). Glossary of transcript symbols with an introduction. Dans G. H. Lerner (Éd.), Conversation analysis: studies from the first generation (p. 13‑ 31). Amsterdam King, D. W., Taft, C., King, L. A., Hammond, C., & Stone, E. R. (2006). Directionality of the association between social support and Posttraumatic Stress Disorder: a longitudinal investigation 1. Journal of Applied Social Psychology, 36(12), 2980-2992.

Page 30 of 32

31

Jo u

rn

al

Pr

e-

pr

oo

f

Laffaye, C., Cavella, S., Drescher, K., & Rosen, C. (2008). Relationships among PTSD symptoms, social support, and support source in veterans with chronic PTSD. Journal of Traumatic Stress, 21(4), 394. Leffler, C., & Dembert, M. (1998). Posttraumatic stress symptoms among U.S. Navy divers recovering TWA Flight 800. Journal of Nervous and Mental Disorders, 186, 574–577 Lehoux, J., Guay, S., Chartrand, É., & Julien, D. (2007). Évaluation préliminaire des qualités psychométriques d'un système de codification du soutien conjugal pour le trouble de stress post-traumatique. Canadian Journal of Behavioural Science/Revue canadienne des sciences du comportement, 39(4), 307. Marshall, E. M., & Frazier, P. A. (2018). Understanding posttrauma reactions within an attachment theory framework. Current opinion in psychology, 25, 167-171. Monson, C. M., Schnurr, P. P., Stevens, S. P., & Guthrie, K. A. (2004). Cognitive–behavioral couple's treatment for posttraumatic stress disorder: Initial findings. Journal of Traumatic Stress, 17(4), 341-344. Nelson Goff, B. S., & Smith, D. B. (2005). Systemic traumatic stress: The couple adaptation to traumatic stress model. Journal of Marital and Family Therapy, 31, 145−157. Nickerson, A., Creamer, M., Forbes, D., McFarlane, A. C., O'donnell, M. L., Silove, D., Steel, Z., Felmingham, K., Hadzi-Pavlovic, D. & Bryant, R. A. (2017). The longitudinal relationship between post-traumatic stress disorder and perceived social support in survivors of traumatic injury. Psychological medicine, 47(1), 115-126. Ozer, E.J., Best, S.R., Lipsey, T.L., & Weiss, D.S. (2003). Predictors of post-traumatic stress disorder and symptoms in adults: A meta-analysis. Psychological Bulletin, 129, 52-73. Paillé, P., & Mucchielli, A. (2012). L'analyse qualitative en sciences humaines et sociales. Armand Colin. Paris. Pizzamiglio, M. T., Julien, D., Parent, M. A., & Chartrand, E. (2001). Système de codification d'interaction de soutien (SCIS global). Document inédit, Université du Québec à Montréal. Ray, C. (1992). Positive et negative social support in a chronic illness. Psychological Reports, 71(3, Pt 1), 977-978. doi: 10.2466/pr0.71.7.977-978 Ray, S. L., & Vanstone, M. (2009). The impact of PTSD on veterans’ family relationships: An interpretative phenomenological inquiry. International Journal of Nursing Studies, 46(6), 838-847. doi: 10.1016/j.ijnurstu.2009.01.002 Regehr, C., Hill, J., & Glancy, G. (2000). Individual predictors of traumatic reactions in firefighters. Journal of Nervous and Mental Disease, 188, 333–339. Regehr, C., Hill, J., Knott, T., & Sault, B. (2003). Social support, self‐ efficacy and trauma in new recruits and experienced firefighters. Stress and Health, 19(4), 189-193. Scarpa, A., Haden, S. C., & Hurley, J. (2006). Community Violence Victimization and Symptoms of Posttraumatic Stress Disorder The Moderating Effects of Coping and Social Support. Journal of Interpersonal Violence, 21(4), 446-469. Shnaider, P., Sijercic, I., Wanklyn, S. G., Suvak, M. K., & Monson, C. M. (2017). The role of social support in cognitive-behavioral conjoint therapy for posttraumatic stress disorder. Behavior therapy, 48(3), 285-294. Solomon, Zahava, Gelkopf, Marc, Bleich, & Avraham (2005). Is terror gender-blind? Gender differences in reaction to terror events. Social psychiatry and psychiatric epidemiology, 40 (12), 947-954. Stanley, I. H., Hom, M. A., Chu, C., Dougherty, S. P., Gallyer, A. J., Spencer-Thomas, S., ... & Sachs-Ericsson, N. J. (2018). Perceptions of belongingness and social support attenuate

Page 31 of 32

32

oo

f

PTSD symptom severity among firefighters: A multistudy investigation. Psychological services. Advance online publication. St-Jean-Trudel, É., Guay, S., Marchand, A., & O’Connor, K. (2005). Développement et validation d’un questionnaire mesurant le soutien social en situation d’anxiété auprès d’une population universitaire. Santé mentale au Québec, 30(2), 43-60. Vogt, D. S., Shipherd, J. C., & Resick, P. A. (2012). Posttraumatic maladaptive beliefs scale: evolution of the personal beliefs and reactions scale. Assessment, 19(3), 308-317. Weiss, D., Marmar, C., Metzler, T., & Ronfeldt, H. (1995). Predicting symptomatic distress in emergency services personnel. Journal of Consulting and Clinical Psychology, 63, 361– 368 Zwiebach, L., Rhodes, J., & Roemer, L. (2010). Resource loss, resource gain, et mental health among survivors of Hurricane Katrina. Journal of Traumatic Stress, 23(6), 751-758. doi: 10.1002/jts.20579

Jo u

rn

al

Pr

e-

pr

Zoellner, L. A., Foa, E. B., & Brigidi, B. D. (1999). Interpersonal friction et PTSD in female victims of sexual et nonsexual assault. Journal of Traumatic Stress, 12(4), 689-700. doi: 10.1023/a:1024777303848

Page 32 of 32